EUVIMEDAtrasti EUVIMED
Scoping ReviewOpen access

Bias Mitigation in Primary Health Care Artificial Intelligence Models: Scoping Review

Bias-Minderung in KI-Modellen der Primärversorgung: Scoping Review

Maxime Sasseville iD, Steven Ouellet iD, Caroline Rhéaume iD, Malek Sahlia iD, Vincent Couture iD, Philippe Després iD, Jean-Sébastien Paquette iD, David Darmon iD, Frédéric Bergeron iD, Marie-Pierre Gagnon iD

Journal of Medical Internet Research · 2025-01-07 · 27 · e60269

DOI: 10.2196/60269 ↗

Auf einen Blick

Die Scoping Review beschreibt Strategien zur Bias-Minderung in KI-Modellen der Primärversorgung. Am häufigsten wurden Ansätze zu Ethnie/Rasse und Geschlecht untersucht; einige Methoden zeigten Potenzial, verschlechterten aber teils die Vorhersage.

Fragestellung

Welche Strategien, Methoden und Ergebnisse zur Bias-Minderung in KI-Modellen der Primärversorgung wurden beschrieben?

Methode

Scoping Review nach JBI; Suche in Medline, CINAHL, PsycINFO und Web of Science für Studien vom 1. Januar 2017 bis 15. November 2022.

Ergebnisse

17 Studien. Häufigste Schutzmerkmale: Rasse/Ethnie in 12 Studien, Sex/Geschlecht in 10. Potenzial zeigten u. a. Relabeling/Reweighing, NLP aus Notizen, Group Recalibration und Equalized Odds.

Einschränkungen

Einige Ansätze erhöhten Vorhersagefehler zwischen Gruppen oder führten zu Fehlkalibrierungen des Modells.

Schlussfolgerung der Autor:innen

Die Autor:innen schließen, dass Bias gegenüber diversen Gruppen leichter mit offenen Daten, mehreren Stakeholdern und in der Preprocessing-Phase gemindert werden kann.

Kernaussagen

  • 17 Studien in der Scoping Review
  • Rasse/Ethnie und Sex/Geschlecht am häufigsten untersucht
  • Einige Fairness-Methoden verbesserten Bias, verschlechterten aber teils die Kalibrierung

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

At a glance

Scoping Review in Journal of Medical Internet Research, published online on 2025-01-07. Written by Maxime Sasseville and 9 co-authors. Institution: Faculté des sciences infirmières, Université Laval, Québec, QC, Canada. Major MEDLINE topics: Artificial Intelligence, Primary Health Care. Cited 38 times so far (Crossref); 47 references. The full text is freely available in PubMed Central.

Auf einen BlickSantraukaSubjectsBibliografiniai duomenysCiteRelated studies

Abstract (original language)

Background

Artificial intelligence (AI) predictive models in primary health care have the potential to enhance population health by rapidly and accurately identifying individuals who should receive care and health services. However, these models also carry the risk of perpetuating or amplifying existing biases toward diverse groups. We identified a gap in the current understanding of strategies used to assess and mitigate bias in primary health care algorithms related to individuals’ personal or protected attributes.

Objective

This study aimed to describe the attempts, strategies, and methods used to mitigate bias in AI models within primary health care, to identify the diverse groups or protected attributes considered, and to evaluate the results of these approaches on both bias reduction and AI model performance.

Methods

We conducted a scoping review following Joanna Briggs Institute (JBI) guidelines, searching Medline (Ovid), CINAHL (EBSCO), PsycINFO (Ovid), and Web of Science databases for studies published between January 1, 2017, and November 15, 2022. Pairs of reviewers independently screened titles and abstracts, applied selection criteria, and performed full-text screening. Discrepancies regarding study inclusion were resolved by consensus. Following reporting standards for AI in health care, we extracted data on study objectives, model features, targeted diverse groups, mitigation strategies used, and results. Using the mixed methods appraisal tool, we appraised the quality of the studies.

Results

After removing 585 duplicates, we screened 1018 titles and abstracts. From the remaining 189 full-text articles, we included 17 studies. The most frequently investigated protected attributes were race (or ethnicity), examined in 12 of the 17 studies, and sex (often identified as gender), typically classified as “male versus female” in 10 of the studies. We categorized bias mitigation approaches into four clusters: (1) modifying existing AI models or datasets, (2) sourcing data from electronic health records, (3) developing tools with a “human-in-the-loop” approach, and (4) identifying ethical principles for informed decision-making. Algorithmic preprocessing methods, such as relabeling and reweighing data, along with natural language processing techniques that extract data from unstructured notes, showed the greatest potential for bias mitigation. Other methods aimed at enhancing model fairness included group recalibration and the application of the equalized odds metric. However, these approaches sometimes exacerbated prediction errors across groups or led to overall model miscalibrations.

Conclusions

The results suggest that biases toward diverse groups are more easily mitigated when data are open-sourced, multiple stakeholders are engaged, and during the algorithm’s preprocessing stage. Further empirical studies that include a broader range of groups, such as Indigenous peoples in Canada, are needed to validate and expand upon these findings. Trial Registration OSF Registry osf.io/9ngz5/; https://osf.io/9ngz5/ International Registered Report Identifier (IRRID) RR2-10.2196/46684

Subjects

MeSH terms (MEDLINE)

Artificial IntelligencePrimary Health CareHumansBiasAlgorithmsFemaleMale

Author keywords

AIalgorithmsartificial intelligencebiascommunity health servicesdecision supportexpert systemhealth disparitiesprimary health carescoping reviewsocial equity

Authors and institutions

  • Faculté des sciences infirmières, Université Laval, Québec, QC, Canada.
  • Vitam Research Center on Sustainable Health, Québec, QC, Canada.
  • Département de médecine familiale et de médecine d'urgence de la Faculté de médecine, Université Laval, Québec, QC, Canada.
  • Research Center of Quebec Heart and Lungs Institute, Québec, QC, Canada.
  • École Nationale des Sciences de l'Informatique, Université de La Manouba, La Manouba, Tunisia.
  • Département de physique, de génie physique et d'optique de la Faculté des sciences et de génie, Université Laval, Québec, QC, Canada.

ORCID profiles: Maxime Sasseville, Steven Ouellet, Caroline Rhéaume, Malek Sahlia, Vincent Couture, Philippe Després, Jean-Sébastien Paquette, David Darmon, Frédéric Bergeron, Marie-Pierre Gagnon

Cite

APA

Sasseville, M., Ouellet, S., Rhéaume, C., Sahlia, M., Couture, V., Després, P., Paquette, J. S., Darmon, D., Bergeron, F., & Gagnon, M. P. (2025). Bias Mitigation in Primary Health Care Artificial Intelligence Models: Scoping Review. Journal of Medical Internet Research, 27, e60269. https://doi.org/10.2196/60269

Vancouver

Sasseville M, Ouellet S, Rhéaume C, Sahlia M, Couture V, Després P, et al. Bias Mitigation in Primary Health Care Artificial Intelligence Models: Scoping Review. Journal of Medical Internet Research. 2025;27:e60269. doi:10.2196/60269

BibTeX
@article{Sasseville2025,
  author = {Sasseville, Maxime and Ouellet, Steven and Rhéaume, Caroline and Sahlia, Malek and Couture, Vincent and Després, Philippe and Paquette, Jean-Sébastien and Darmon, David and Bergeron, Frédéric and Gagnon, Marie-Pierre},
  title = {Bias Mitigation in Primary Health Care Artificial Intelligence Models: Scoping Review},
  journal = {Journal of Medical Internet Research},
  year = {2025},
  volume = {27},
  pages = {e60269},
  doi = {10.2196/60269}
}
Start for free →